Lattice
Spatial proteomics from H&E
Lattice predicts spatially resolved protein-marker maps from routine H&E, delivered as virtual mIF channels. No new assay, no new hardware, and no tissue consumed.
One slide in. A stack of protein maps out.
Lattice takes a standard H&E image and predicts spatially resolved protein expression across a broad panel of immune, structural and functional markers. The output sits at the resolution of the underlying image, so you can work at whole-slide scale or zoom to individual cells.
Pan-cancer
Built across a wide span of tumour and tissue types rather than a single indication with a fixed marker panel.
Cohort scale
Runs across an entire archive, so a retrospective study is a compute question rather than a tissue question.
Tunable to you
Fine-tunes to your indications and your target panel when the question is specific enough to need it.
The tissue is the bottleneck.
Multiplex immunofluorescence is expensive, slow, and consumes the sample. Biopsy requirements in trials keep tightening, most archival material is scarce or exhausted, and re-acquisition is usually impossible. So the spatial layer that would answer the question is missing from almost every cohort that matters.
H&E is the one thing everybody already has. Lattice predicts spatial protein maps from it, which puts decades of shelved slides back in play.
Delivered as a service, not as homework.
We run Lattice end to end. Your team sends slides and gets answers, with nothing to install, validate or staff.
- 01
Send H&E
Digitised slides from a trial cohort or an archive, with whatever clinical annotation you can share.
- 02
We run and analyse
Predicted spatial proteomics across the cohort, then segmentation, phenotyping and the spatial analysis on top of it.
- 03
Get a decision-ready readout
Structured spatial data plus a written analysis aimed at the biomarker question you actually asked.
For cohorts that cannot leave your institution, we deploy and run Lattice entirely within your environment.
Where teams point it
Target screening
Look at target expression and its spatial context across cohorts where the assay was never run.
Patient stratification
Group a cohort by spatial phenotype and see how the tumour microenvironment differs between the groups.
Retrospective rescue
Profile archives that were never intended for spatial work, without consuming a single section.
Wet-lab triage
Decide which samples justify the cost of real multiplex work instead of guessing at the shortlist.
Bring us a cohort.
Tell us the biology you are chasing and what slides you have. We will tell you honestly whether Lattice is the right instrument for it.
Research use only (RUO). Not for use in clinical or diagnostic procedures.